The adaptive immune system has evolved several different lines of attack, each one optimally effective against a given pathogen. For example, in response to intracellular microbes, CD4+ T-helper cells (Th) differentiate into Th1 cells; in contrast helminths induce Th2 cells, whose cytokines (IL-4, IL-5 & IL-10) induce IgE and eosinophil-mediated destruction of the pathogens. Although these diverse responses have been characterized in detail, the mechanism by which a particular type of immune response is initiated is poorly understood. Two recent studies suggest a possible mechanism: (i) Distinct types of dendritic cells can initiate different Th responses. In mice, lymphoid DCs initiate Th1 responses, while myeloid DCs initiate Th2 responses (Pulendran et al, 1999; Maldonado et al, 1999). (ii) Different microbial antigens signal through distinct pattern recognition receptors on antigen-presenting cells. For example, LPS from E. coli signals through the Toll-like receptor 4 (TLR4), while cell wall components from Gram positive bacteria signal through TLR2. Here we propose that, analogous to the situation in Drosophila, signaling through distinct TLRs can yield qualitatively different immune responses. Furthermore, we suggest that different DC subsets are endowed with a distinct, repertoires of pattern recognition receptors, which enable them to recognize distinct classes of pathogens, and initiate different types of immune responses. In this proposal, we will co-inject a soluble protein (ovalbumin) with either of two different LPs molecules that signal through distinct pattern recognition receptors : (i) E. coli LPS, which signals through TLR4, and induces Th1 cytokines, and (ii) P. gingivalis LPS, which signals independently of TLR4. P. gingivalis infections are often characterized by Th2 cytokines. Using this approach, our Aims are: (1) To determine the type of antigen-specific CD4+ and CD8+ T cell responses against soluble ovalbumin co-injected with either LPS; (2) To determine the effect of either LPS on various DC subsets, in vivo and in vitro; (3) To determine whether TLR2 is involved in P. gingivalis signaling; (4) To determine whether injecting Lipid A from P. gingivalis can redirect autoimmune Th1 responses towards benign Th2 responses. This research will provide us with a novel mechanism by which distinct pathogens can elicit distinct adaptive immunity, by targeting different cells of the innate immune system. It should also offer novel strategies for manipulating immune responses in clinical settings.